An automated backup and recovery utility that synchronizes OpenClaw configurations, workspace data, and installed skills with a Git repository.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clone-lobster
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install clone-lobster using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Clone Lobster?
Clone Lobster is a specialized lifecycle management tool designed to ensure the persistence and portability of your AI agent environment. It serves as a bridge between your local OpenClaw setup and remote version control, capturing everything from core workspace documents like AGENTS.md and SOUL.md to system-level supervisor configurations. By utilizing this tool, developers can maintain a consistent state across different machines and safeguard their custom Openclaw Skills against data loss or corruption.
The skill operates by monitoring critical directories and triggering synchronization events based on specific user actions or environment changes. Whether you are modifying agent personalities, installing new capabilities, or updating system-wide scripts, Clone Lobster ensures that your entire technical context is versioned and recoverable through a standard Git workflow.
Clone Lobster Use Cases
- Automatically protecting workspace files like AGENTS.md, SOUL.md, and USER.md after significant updates.
- Synchronizing your library of Openclaw Skills across multiple development environments.
- Creating a recovery point before making experimental changes to the openclaw.json configuration.
- Persisting session databases and memory indices to maintain long-term context across agent restarts.
- Restoring specific environment components, such as only the skills or only the workspace files, to a fresh installation.
How Clone Lobster Works
- The system identifies changes in the local environment, such as file modifications in the workspace or the addition of new Openclaw Skills.
- A backup script is triggered either manually by the user or automatically by the AI agent upon detecting significant configuration changes.
- The tool executes a Git-based synchronization process that pushes the current state of selected directories to a remote repository defined by the CLONE_LOBSTER_REPO_URL.
- During the backup, the tool intelligently filters out redundant data like node_modules or python cache files to keep the repository clean.
- When recovery is needed, the restore script pulls the specified version from the Git repository and re-populates the local filesystem to match the saved state.
Clone Lobster Setup
To initialize the backup system, you must have SSH access to your Git repository and set the environment variable for your target repository URL.
# Set the repository URL
export CLONE_LOBSTER_REPO_URL="ssh://[email protected]/your-repo.git"
# Perform an initial manual backup
bash scripts/backup.sh "Initial snapshot of my Openclaw Skills environment"
To restore your environment from a previous backup:
# Restore the entire environment
bash scripts/restore.sh --all
# Or restore specific components
bash scripts/restore.sh --skills
bash scripts/restore.sh --workspace
Clone Lobster Data Schema & Taxonomy
| Category | File Path | Data Type |
|---|---|---|
| Workspace | ~/.openclaw/workspace/ |
Agent definitions, memories, and user profiles |
| Configuration | ~/.openclaw/ |
Main app settings and execution approvals |
| Openclaw Skills | ~/.openclaw/skills/ |
Source code and assets for installed agent skills |
| System | /etc/supervisor/ |
Process management and startup configurations |
| Context | ~/.openclaw/data/ |
SQLite session databases and memory indices |
Clone Lobster Advanced Features
- Intelligent auto-trigger rules that execute backups at the end of high-impact conversations.
- Granular restoration flags that allow for selective recovery of skills, configurations, or workspace data.
- Automatic exclusion of build artifacts and temporary files using internal filter logic.
- Idempotent script execution, ensuring that repeated backup or restore commands do not corrupt the environment state.
- Support for periodic snapshots via integration with system heartbeat checks.
SKILL.md
Loading
Related Openclaw Skills
Remote Job Finder
A conversational AI agent skill that integrates Remote Rocketship with WhatsApp to help users find and monitor remote job listings through natural language.
Language Learning Tutor
An expert AI polyglot tutor that provides adaptive, conversational language instruction and structured drills for over 100 languages.
Humanizer
A sophisticated text editing utility that strips away recognizable AI-generated writing patterns to produce natural, human-sounding content.
Gemini Deep Research
A powerful AI agent skill that performs systematic web searches and synthesizes multi-source data into comprehensive reports.
CrabPet - AI Lobster Companion
A virtual pixel-art lobster that evolves and levels up based on your activity and engagement within Openclaw Skills.
Railil
A CLI-based Israel Rail schedule search tool with fuzzy station matching and multi-format output.
Featured*
